News organizations, including The New York Times, are set to begin searching ChatGPT logs after a court ruling against OpenAI. The data will be accessed through agreed-upon keywords, and although it may help illustrate market dilution, it's not considered essential for the plaintiffs' cases. Concerns have been raised about data security during the process due to the sensitive nature of the information involved.

The Japanese government has requested OpenAI to avoid copyright infringement concerning its new video generation app, Sora 2, which generates videos featuring copyrighted anime and game characters. Officials, including Minoru Kiuchi, emphasized the importance of protecting Japan's intellectual property, highlighting the potential implications of AI technologies on creative works while expressing hopes for voluntary compliance from OpenAI. Concerns about copyright violations have sparked significant backlash on social media and raised discussions on the need for clearer AI regulations.

OpenAI is contesting a court order that requires the preservation of all ChatGPT user logs, including deleted and sensitive data, amid accusations from news organizations regarding copyright infringements. The company argues that the order jeopardizes user privacy and was issued without sufficient evidence of necessity, as it stems from concerns about users potentially deleting chats to hide their activity. OpenAI seeks to have the order vacated until a substantial need for such preservation can be demonstrated.

OpenAI's new video generator, Sora, will mandate that copyright holders opt out if they do not wish their content to be used. This policy reflects a shift in how AI companies are navigating copyright issues, while ensuring that recognizable public figures' likenesses are not generated without permission.

OpenAI is set to launch a TikTok-like social app powered by its Sora 2 video model, which will exclusively feature AI-generated content and restrict user uploads from their camera rolls. The app will allow clips up to 10 seconds long and include identity verification tools for users to control the use of their likeness in generated videos. Copyright protections are planned, but their effectiveness remains uncertain as rights holders will need to opt-out for their content to be excluded.